Hybridization probes used in DNA and RNA assays are labeled in some fashion to facilitate detection of the duplex after the probe has hybridized with a complementary strand of "target" DNA or RNA in the sample under analysis. Most commonly, the probe can be labeled by enzymatically incorporating radio-labeled nucleotides on the 3' or 5' terminus of the probe. [A. M. Maxam et al., Meth. Enzymol. Volume 65, 499-560 (1980)]. Alternatively, higher levels of radiolabeled nucleotides can be incorporated by nick translation [P. W. J. Rigby et al., J. Mol. Biol. Volume 113, 237-251 (1977)]. This latter method possesses the inherent advantage of being able to incorporate one radioactive phosphorus per constituent nucleotide in the probe and thus assays using these probes are characterized by the most sensitive detection limits in hybridization assays. The obvious disadvantages of these probes are the hazards and inconveniences associated with the use of radioisotopes.
A second type of label involves enzymatically synthesizing a probe with a mixture of nucleotides containing a biotinylated pyrimidine base (uracil or adenosine) incorporated into one of the nucleotides [e.g., P. R. Langer et al., Proc. Natl. Acad. Sci. USA Volume 78, 6633-6637 (1981)]. Alternatively, biotinylated histone Hl proteins can be chemically crosslinked with bases in DNA to yield a labeled probe [M. Renz, EMBO J. Volume 2(6), 817-822 (1983)]. After hybridization of the probe to a complementary strand of target DNA, the hybrid is treated with a reporter molecule attached to avidin which binds tenaciously to the biotin moieties in the probe. The reporter can be either an opaque polymeric microsphere bound to avidin [O. C. Richards et al., Proc. Natl. Acad. Sci. USA Volume 76(2), 676-680 (1979)] or avidin-ferritin [A. Sodja et al., Nucl. Acids Res. Volume 5(2), 385-401 (1978)]. Each of these can be visualized via electron microscopy. Alternatively, the reporter may be an avidin-enzyme complex which when treated with the enzyme's substrate, will yield a visually detectable colored product [J. J. Leary et al., Proc. Natl. Acad. Sci. USA, Volume 80 4045-4049 (1983)]. Assays based on these probes obviate the use of radioactive materials; however, only about 2% of the nucleotides in a given probe molecule can be derivatized with a biotin moiety without reducing the specificity of the probe for its complementary strand of sample DNA. This reduction in specificity arises from modification of the ability of complementary bases to pair properly because of the presence of the biotin moiety. The reduced incorporation of label in these probes relative to radioactive probes results in a poorer detection sensitivity.
A third type of label is an enzyme directly attached to the probe such that after hybridization the enzyme in the hybrid is treated with substrate to yield a colored product. The enzyme can be bonded directly to bases in the probe [M. Renz et al., Nucl. Acids Res., Volume 12(8), 3435-3444 (1984)] or to bases in another strand of DNA complementary to a length of oligonucleotide chemically attached to the end of the probe [J. G. Woodhead et al., Biochem. Soc., Trans., Volume 12(2), 279-280 (1984)]. In the latter case, the probe is hybridized to complementary target DNA and this hybrid is treated with the oligonucleotide containing the enzyme labels. This strand of labeled oligonucleotide binds with the single-stranded tail of complementary DNA which had been chemically attached to the end of the probe. Visualization of the enzyme reporters is accomplished by conversion of the enzyme's substrate to colored product. The primary difficulties associated with this procedure arise from the low level of enzyme attachment and the loss of enzyme activity when subjected to the stringent conditions (e.g. elevated temperature, nonaqueous solvent) typically used in the hybridization protocol. A secondary difficulty associated with the latter of these labeling methods is the possibility that the synthetic strand of oligonucleotide containing the enzyme label might be complementary to a sequence of naturally-occurring nucleotides in the sample. Thus, in addition to pairing with the sequence of bases attached to the end of the hybridization probe, the enzyme-labeled oligonucleotide may bind nonspecifically to natural sequences of complementary bases and thus lead to background problems in samples used as negative controls or blanks.
A fourth type of label involves coupling a fluorophore either to bases in the constituent nucleotides of the probe [C. H. Yang et al., J. Biochem. Volume 13, 3615-3620 (1974)] or to the 3'-terminus [R. W. Richardson et al., Nucl. Acids Res. Volume 11(8), 6167-6184 (1983)] or to the 5'=terminus [C. H. Yang et al., Arch. Biochem. Biophys., Volume 155, 70-81 (1973)] of the probe. As above, the sensitivity of these methods is limited by the small number of labels which can be incorporated into each copy of the probe.
A fifth type of label involves the use of an antibody directed against some antigenic determinant in the probe or in the target-probe duplex. In the former case, an antigenic determinant (e.g., biotin, bromine, N-acetoxy-N-2-acetylaminofluorene) is covalently coupled to bases in the nucleotide of the probe [e.g., L. Manuelidis et al., J. Cell Biol., Volume 95, 617-625 (1982)]. These antigenic determinants may interfere with specific hybridization and limit the utility of the probe. In the latter case, the double-stranded DNA-RNA hybrid itself is immunologically distinguishable from DNA-DNA and RNA-RNA duplexes [G. T. Rudkin et al., Nature, Volume 265, 472-473 (1977)]. The antigenic determinant in the probe-target duplex was detected with an antibodyfluorophore conjugate and fluorescence microscopy. The labeled probe can also be detected with antibodies to which an enzyme (L. Manuelidis et al., op. cit.) or colloidal gold [N. J. Hutchinson et al., J. Cell Biol., Volume 95, 609-618 (1982)] has been conjugated. Again, the ultimate sensitivity is limited by the number of labels available to detect a hybridization event.
The instant invention overcomes the limitation of the prior art by incorporating the maximal number of nonradioactive labels, one per nucleotide, into each probe with negligible effect on hybridization efficiency or specificity.